Menu
Back previous/next report: k j

Intel 1-node-2S-GNR_128C

Submitted by Intelon 2026-06-26

Submission Info
Submitter Intel
Submission ID f82a98ec-b5f8-42ea-b1c4-ded6f25a822b
Submission Status PUBLISHED
Submission Date 2026-06-26
Publication Date
Division Standardized
Model
Model Name Llama 3 8B
Division Standardized
Model Precision FP16
Calibration Link GitHub
Model Notes W4A8 quantization with vLLM's quantization tools, with fp8 kv cache. No other model transformations applied.
Datasets
Dataset Name CNN/DailyMail
Dataset Type performance
Average Input Tokens 870
Dataset Link Hugging Face

System TPS vs. Tokens/s per User

05001.0k1.5kSystem TPS0102030Tokens/s per User

System TPS vs. Concurrency

05001.0k1.5kSystem TPS01,0002,0003,000# of queries

Time to First Token P95 vs. Concurrency

050100150TTFT P95 (s)01,0002,0003,000# of queries

Tokens/s per User vs. Concurrency

0102030Tokens/s per User01,0002,0003,000# of queries
SUT Summary
Intel
System Name 1-node-2S-GNR_128C
System Availability Available
System Category Datacenter · —
System Size 1x2S-GNR-AP
Node Types 1
Total Node Count 1
Max Supported Concurrency 3,072
Node Type ID 0
Node Count 1
Software
Framework vLLM 0.23.0
Serving Framework vLLM 0.23.0
Inference Backend Not detected: CUDA/ROCm/XPU runtime not found
Driver Not available
Link to Container
Operating System centos 9
Filesystem ext4 iso9660 ramfs vfat xfs
Other Software
Software Notes
Hardware

Processor

Processor Model Name Intel(R) Xeon(R) 6980P
Processors per Node 2
Cores Per Processor 256
VCPUs Per Processor 256

Accelerator

Accelerator Model Name Not available
Accelerators per Node 0
Memory Type Not available
Memory Capacity Not available
Accelerator Interconnect Not available
Host-Accelerator Interconnect Not available

Host / Storage

Host Memory Capacity 2.3T
Memory Configuration 2304GB (24x96GB DDR5 8800 MT/s)
Storage Capacity 4.9 TB NVMe SSD, 243 GB SSD
Storage Type NVMe SSD

Networking

Network Interfaces 1x Ethernet
Network Details Ethernet
Network Topology

Other

Other Hardware
Cooling
Hardware Notes
Run Data
Field NameRun 1Run 2Run 3Run 4Run 5Run 6Run 7Run 8Run 9Run 10
Run Date06/26/202606/26/202606/26/202606/26/202606/26/202606/26/202606/26/202606/26/202606/26/202606/26/2026
Concurrency316401282565121,0241,5362,0483,072
System Tokens/Second96.5366.2648.81,045.11,306.11,479.31,607.41,636.31,742.21,767.3
Tokens/Second per User32.222.916.28.25.12.91.61.10.90.6
TTFT P95 (ms)520.7619.7687.31358.51910.93941.527324.955763.381136.7139534.3
Utilization5.5%20.7%36.7%59.1%73.9%83.7%91.0%92.6%98.6%100.0%
Configuration Summaryllama-3.1-8b-concurrency · submission · c3llama-3.1-8b-concurrency · submission · c16llama-3.1-8b-concurrency · submission · c40llama-3.1-8b-concurrency · submission · c128llama-3.1-8b-concurrency · submission · c256llama-3.1-8b-concurrency · submission · c512llama-3.1-8b-concurrency · submission · c1024llama-3.1-8b-concurrency · submission · c1536llama-3.1-8b-concurrency · submission · c2048llama-3.1-8b-concurrency · submission · c3072
Config Typesubmissionsubmissionsubmissionsubmissionsubmissionsubmissionsubmissionsubmissionsubmissionsubmission
Config Region
Target QPS
Result Statecompletecompletecompletecompletecompletecompletecompletecompletecompletecomplete
Measured Accuracy Score
Benchmark Versionbbc8697bbc8697bbc8697bbc8697c81001bc81001bc81001bbbc8697c81001bc81001b
Result Version0.1.00.1.00.1.00.1.00.1.00.1.00.1.00.1.00.1.00.1.0
Archive URIgs://mlperf-endpoints-blob/runs/f2ed1200-7d9f-4a09-b222-0d0b6533c440/archive.tar.gzgs://mlperf-endpoints-blob/runs/fb68bc14-e88f-4803-8d67-9ea61be1544a/archive.tar.gzgs://mlperf-endpoints-blob/runs/802201ce-3015-4c24-8401-37b1b8025048/archive.tar.gzgs://mlperf-endpoints-blob/runs/08cd248c-087f-486e-b173-d58df0a1ff3c/archive.tar.gzgs://mlperf-endpoints-blob/runs/72e238f4-b53c-428d-8b53-1b6a9da0b690/archive.tar.gzgs://mlperf-endpoints-blob/runs/4b98ac30-88a4-41e3-ba5e-c8a1abc6fd27/archive.tar.gzgs://mlperf-endpoints-blob/runs/f8b8081e-49f6-4a84-a1eb-6cc0b084dd63/archive.tar.gzgs://mlperf-endpoints-blob/runs/d16a14db-3524-4281-aca7-02e158498689/archive.tar.gzgs://mlperf-endpoints-blob/runs/91dcd2dd-996a-4629-8280-245790ecfb17/archive.tar.gzgs://mlperf-endpoints-blob/runs/dbef15d2-82c9-4546-b316-09e2b8d7acd3/archive.tar.gz
QPS0.75412.86085.06908.164510.203611.557112.557812.783513.610913.8067
Total Output Tokens1,711,1031,711,1011,711,1011,711,1041,711,1021,711,1021,711,0991,711,1021,711,1021,711,103
Average Output Tokens128128128128128128128128128128
Run Duration (s)17,726.314,672.832,637.191,637.331,310.121,156.691,064.521,045.72982.15968.22
Total Queries13,36813,36813,36813,36813,36813,36813,36813,36813,36813,368
Total Queries Issued13,36813,36813,36813,36813,36813,36813,36813,36813,36813,368
Total Queries Failed0000000000
Time to First Token (TTFT) P95 (ms)
Minimum85.989.5105.1135.5169.3227.8240.0225.1233.2267.1
Average285.8337.1368.4702.71051.91817.24190.97885.212236.424437.1
P50255.9295.8313.3554.9711.1945.01410.41874.42473.73603.5
P90453.6543.4589.21119.91493.72064.14583.214332.545042.0106156.8
P95520.7619.7687.31358.51910.93941.527324.955763.381136.7139534.3
P99612.4822.11007.63005.813652.027193.554954.283933.3115867.3178139.4
P999733.01217.73214.97638.014406.930214.361711.193520.8132293.4191547.7
Maximum1142.31654.73289.87641.815308.330254.761722.093532.6132310.6191557.9
Time Per Output Token (TPOT) (ms)
Minimum26.230.633.353.360.066.1110.7109.8163.2289.3
Average29.141.459.2117.6188.6332.1601.0861.61068.91514.7
P5028.840.958.6116.5181.6305.9566.4711.81017.91437.8
P9031.746.568.3141.4251.7489.6925.21881.41656.12448.4
P9532.548.471.0149.1278.9547.61040.52048.21977.32616.5
P9933.953.177.7161.9335.7615.51275.32300.82290.23033.7
P99935.459.791.4179.4372.6687.01297.62348.32335.53286.4
Maximum36.563.898.4190.8393.4696.51301.42363.32423.93373.4
Query Latency (ms)
Minimum3434.24046.04451.87104.28361.79827.314742.314452.122115.438962.8
Average3977.55589.67884.915638.025000.344000.280523.0117302.4147992.8216804.5
P503952.85532.07806.515472.224073.840781.774850.694636.2136765.5192964.3
P904358.96271.29067.918769.533068.665139.6129268.5245693.1237623.2355790.0
P954474.76530.19440.419734.636730.970851.1140274.1265098.1275355.0418511.6
P994689.57158.710288.621581.043729.979373.4164225.2296386.9326615.5467813.3
P9994930.67953.612029.723503.148166.688689.2167302.5301168.8347126.8479231.4
Maximum5173.98470.513117.325038.550329.789355.8168831.7302755.7347140.4479234.6